0:00
We're continuing and our next speaker is Richard Broadhead from Quantum Si.
0:05
Richard, he's a senior seasoned expert in proteomics with nearly three decades of experiment experience spending academia and industry as a field application scientist for Quantum Si he provides expert scientific, technical and application support to researchers.
0:24
His role is pivotal in ensuring that the users maximise the value of Quantum Si’s technology by guiding experimental design, implementation, data analysis and interpretation.
0:36
Dr Broadhead is deeply committed to advancing the role of proteomics in biomedicine, leveraging his extensive experience to drive innovation in drug discovery and development.
0:45
With without further ado, the floor is yours.
0:54
Yep, Thank you very much.
0:55
And thanks everyone here for coming along.
0:58
So very much appreciate everyone taking the time to come here.
1:03
So yeah, my name is Richard, just based in Oxford, just up the road.
1:07
And I'm here to talk about the protein sequencing platform technology from Quantum Si.
1:33
So what I'm going to do is just give an overview of the some of the applications, talk a little bit about the technology and hopefully give you a feel as to how it can be used and be very happy to discuss any additional applications that you might have a little bit further on.
1:53
So I mean, this slide just gives an overview of some of the types of applications that the system has been used for.
2:00
I mean, again, at the heart of the system, it is a protein sequencer.
2:04
But beyond that, there are a wide number of other potential applications and other biological insight areas that the system could be applied to.
2:16
Now, if you've been past our booth, you've probably seen the instrument that we call the Platinum Pro.
2:22
Now, ultimately this is designed to sequence proteins, single amino acid resolution.
2:28
And I will talk about a little bit about how that works just in a moment.
2:32
And again, it is actually the bench top system.
2:36
So it will actually sit on the bench.
2:38
It's about the same size as the PCR machines, so very compact and very accessible as well.
2:44
And it does actually directly detect the proteins through direct interaction with the amino acids.
2:50
Now we'll describe now a bit about how it works.
2:54
So it's essentially a three-stage process.
2:57
So the first stage is, as you might expect, sample prep.
3:01
And we actually supply everything needed for this, which is quite similar to something you'd be familiar with from our spec workflow where you would digest your proteins.
3:10
We use Lys-C for reasons that we'll come onto in a moment.
3:14
And then once the proteins have been digested into peptides, those are then loaded onto the microfluidic chip and run on the Platinum Pro machine.
3:23
And this one sequences individual peptides loaded into wells across the chip.
3:28
And as that's happening, the data is uploaded into a cloud based analysis system and then this is then analysed and output as a sequence data.
3:38
But also more detailed raw data is available as well if you want it to interrogate things at that level.
3:46
So just to talk a bit about the sample prep, the first step is to digest the proteins.
3:53
This is done using Lys-C, which cuts that's lysine and functionalises the peptides by creating a C-terminal lysine.
4:03
So then a linker is then added, which at one end links to the C-terminal lysine and then the other binds of peptides onto the chip using a streptavidin biotin reaction.
4:14
So this then loads your chip and again, there's minimal hands-on requirements.
4:21
Basically, if you can hold up your pipette, you can do this.
4:25
So it's very simple, very easy to do.
4:29
And then the heart of the system is what happens on the Platinum Pro instrument itself.
4:34
So what happens is that we add two key classes of molecules.
4:39
So you see on the slide here, you'll see this is an example peptide immobilised into a well.
4:46
And then as the sequencing reaction happens, the first thing is that a molecule we call a recogniser binds to the n-terminal amino acid on the peptide.
4:55
Now these recognisers are specific for particular amino acids.
4:59
So the recogniser will bind to the amino acid and when that recogniser binds you will get a full pulse of fluorescence and the length of time or the duration of that pulse is determined by the identity of the amino acid being sequenced.
5:13
So what you would get is so for example here you get the arginine R sequence, so you get the recogniser binding then being released.
5:21
Then the aminopeptidases comes in cuts the peptide chain that releases the next amino acid, in this case leucine for sequencing.
5:30
And then that generates a fluorescence pulse, which again is recorded on the Platinum instruments.
5:35
And this happens all the way through the all the way through the sequencing reaction.
5:41
So what you have is this happening across 2 million wells massively parallel across the whole chip.
5:46
So all of these peptides have been sequenced and as that sequencing is happening, it's then uploaded into the Platinum analysis software.
5:56
And this is where the actual sequence data or the raw fluorescence data is converted into protein sequence.
6:03
So we call the sort of the fluorescence duration generated the kinetic signature of the protein because the length of time that fluorescence pulse happens for is determined by the kinetics of the on off reaction of the recogniser.
6:21
And that, as I will talk about soon, could also be influenced by other factors such as post translational modifications.
6:27
So can be a factor in detecting those.
6:30
But in terms of how it works, there's no dedicated computer required.
6:35
Everything is done in the cloud.
6:36
So and you can log onto your account for absolutely anywhere.
6:39
So if you're really keen to see how the runs going, you can log on from home.
6:43
So depends on your work life balance.
6:46
But otherwise you can come in the next morning and see how the reaction has gone.
6:53
Now, the sort of data that you will see is presented up on here.
6:58
So what you will see is the protein digested into peptides and each colour associated with each amino acid residue shows within each peptide which recogniser would be involved for sequencing that peptide.
7:15
What you also get reported is a number of times that peptide was sequenced, which is reported as a number of alignments.
7:22
You get to your reference faster file and also, we report a false discovery rate as well.
7:27
So this is calculated using a decoy peptide method in a very similar way to what is used with most mass spec workflows.
7:38
So I mean, that is the very basics of the system and how it works.
7:42
So what I will do now is talk about some example applications and do a bit deeper dive into how the system is able to inform your protein science.
7:55
So the sort of output that you will get, and this is actually images from screenshots of the analysis software, what it will output using this protein IL4 as an example, it will tell you the number of total number of alignments.
8:11
So that's the total number of peptides that you've sequenced.
8:15
It will give you an overall total and a total for each peptide.
8:19
So for example, if you look at on the right, this EAN peptide was sequenced 15,548 times false discovery rate 0.
8:28
So you can be very confident that was there at the top.
8:32
This is also output in the software.
8:34
So this shows which residues were sequence able and the colour code shows the specific recognizer that will have been used in the sequencing reaction whereas the grey ones were not sequenced.
8:48
And then what you can do as well in the software is do a much deeper dive.
8:52
So for example, here you can see for each individual peptide you get a readout showing the coverage.
8:58
So it is showing how many times out of each time that peptide was sequenced that given residue was sequenced.
9:05
And the graphs at the bottom actually show you the pulse duration of the fluorescent signal that was used to call that protein or peptide.
9:14
So this will give you a deep insight both into what you have seen and the data that was used to call that.
9:21
And looking at this pulse duration data in more detail will allow you to look at any changes or potentially PTMs associated with the sequencing of that residue as well.
9:34
And the question that does often come up is how does this compare to mass spec workflows?
9:39
Well, what's very interesting is what we tend to find and again, this preprint shows this quite well, is that the peptides we see are often ones that aren't necessarily seen on mass spec, but also mass spec can sometimes pick up the ones that we don't see as well.
9:54
So it is can actually be very complementary as well as being an easy to use simple bench top system.
10:01
It can actually add additional insights if you already have mass spec capability accessible to yourselves.
10:07
So it can add that additional level of value.
10:11
And again, we can very happily share copies of this preprint if that would be of interest to anyone.
10:18
And something that the system is actually very good at because it does actually interact with the amino acids directly is precise detection of variations on a single amino acid level.
10:31
And this is something where it does actually start to come into its own.
10:35
So for example, if specific changes at a given amino acid are of interest, this is an example where we compared to antibodies and we're looking for a particular PTM or a particular change at the amino acid level.
10:51
And this is where aspartic acid becomes iso-aspartic acid.
10:55
So you can see if you look at the histograms here, you'll see, yeah.
11:01
So you'll see this D residue.
11:03
You will see here the pulse duration for aspartic acid is 0.42 seconds.
11:09
However, when that changes to iso-aspartic acid, you'll see that the pulse duration there changes to 1.27 seconds.
11:17
So you can actually see a distinct change because of this PTM happening at the at this point.
11:24
And again, that is observable because of the direct interaction between the recognizer and the amino acid being sequenced.
11:34
And again, this is something that we have actually published in a paper.
11:38
This is currently on Biorxiv undergoing review.
11:42
This is from Gloria Sheynkman's lab at the University of Virginia in the US and this is some data that we'll talk about a little bit more here.
11:52
Now what they're looking at is tropomyosin and this protein is processed and also spliced at the RNA level to produce different proteoforms.
12:02
And what this lab has been very interested in is saying is looking at these proteoforms and saying can we characterise, can we understand these in more detail?
12:11
So one change that does happen is you have a mutation that substitutes a leucine for an isoleucine.
12:19
Now again, this is where the system can complement mass spec really well because leucine to isoleucine can be very difficult because the two amino acids are essentially isobaric.
12:31
So it can be very tricky to pick up with MS.
12:34
Whereas if you look here, you have a pulse duration of 0.38 seconds with a leucine, 0.24 seconds with the isoleucine.
12:40
So for our system, this is really easy to pick up.
12:44
So this is something that's very simple to do.
12:48
And it's interesting that it is actually the same recognizer that is involved in detecting both leucine and isoleucine.
12:54
But because the kinetics of the on off binding are different, the pulse duration is different.
12:59
So this makes it very easy to see and again, if you do want a copy of the preprint of this article, we're very happy to share it.
13:09
So as well as changes the protein level due to mutations, it is actually, it's actually also subject to splicing.
13:19
So this happens at the RNA level.
13:22
So you'll see here, so for example, you have the two forms of the protein.
13:27
And then you'll see from position amino acid 8 onwards, you'll see there is a change again at the individual amino acid level.
13:35
So here you have LAS, here you have VYA.
13:39
And again, you can see there's a very clear change in the pulse duration, which manifests as a change in the protein sequence.
13:49
So again, that is very easy to pick up with our system.
13:54
And again, this is as it's reported in the software.
13:57
So this is what you see.
13:58
But again, in the software, you can actually do a much deeper dive and look at each peptide that called those amino acids individually if you really want it.
14:08
So this is the sort of detail that you can obtain.
14:16
So far I've been talking about using the system to directly look at proteins and amino acids.
14:24
However, there isn't another workflow that is starting to become sort of much more sort of prominent and people expressing much more interesting it now.
14:33
And this is the protein barcoding workflow.
14:37
Now for barcoding, this is something we do actually have a kit for and this is some this is an application where rather than sequencing your protein directly, you would append a barcode of amino acids.
14:49
So particular amino acid sequence, you would append the codons for that onto the DNA by molecular biology.
14:57
So rather than sequencing the protein of interest directly, what you would do is sequence this barcode, which essentially acts as a proxy for your protein as a reporter sequence.
15:06
So this is a sequence that we know is very easy to pick up on the Platinum system sequences very well.
15:13
So you can use this as a reporter for proteins that might otherwise be difficult to detect or sequence.
15:21
And there are a variety of applications for this.
15:25
For example, you can actually look at just direct detection of a protein as to whether it's expressed.
15:31
If that protein is difficult to prepare or look at again, you can actually use it to check if relative levels of given proteins are interesting because you know the, you know, the sequence ability or the amenability of sequencing for the barcode.
15:50
So this is something that has a multitude of applications.
15:53
And as I say here, these are examples of the barcodes that we have developed.
15:59
So there are 8 shown here.
16:01
More are constantly being developed, but this is ones that are available as part of our kit as well.
16:08
So the first example that I will talk about is using this to look for nano body enrichment to characterise nano bodies and to see whether those nano bodies do actually have the specification that they are intended to.
16:24
Now these proteins can be very difficult to sequence directly, but using the barcodes, you can use them as a proxy to check the specificity of the nano body.
16:35
So what we did is we recombinantly expressed these nano bodies.
16:40
One of these nano bodies was raised against MBP maltose binding protein, the other one against GFP green fluorescent protein.
16:49
So these were just used as standard well known proteins.
16:52
So just as a just conveniently as a model.
16:55
And then to those particular nano bodies, we appended one with one peptide barcode and the other with another peptide barcode.
17:04
So two different peptide barcodes for the two different nano bodies.
17:08
We then run these nano bodies through a GFP coated column.
17:13
So what you would expect is that the anti GFP nanobody would bind and the anti MVP nano body would not.
17:21
But the purpose of this experiment was just to verify that is in fact what was happening.
17:27
So once that had been happening, we then eluted cleaved the barcode from the LU8 and then sequence them on the platinum.
17:38
Now what we found is that you looked at if you looked at the ratio between the two different nano bodies found in the mix before selection, the difference was about 5 to 1.
17:49
So one of the antibodies was more prevalent than the other one in the mixture, but only by a fairly small factor.
17:57
Whereas after running through the column enrichment and then checking the barcodes that were present, you could see that the ratio changed dramatically, showing a dramatic enrichment, showing that the specificity of the nano body was exactly what was expected.
18:17
Now the nano body itself would have been very difficult to sequence directly, but if you look for the barcode that acts as a proxy for the presence of that nano body.
18:25
So you could then infer what the abundance of that nano body was.
18:32
So this was the first application of barcoding that really was sort of thought about a year ago.
18:38
But within the last year, there's been a very interesting novel application which has developed and that is actually using protein barcoding to look at the efficiency of therapeutic mRNA or siRNA delivery using lipid nanoparticles, which has become well, much more well known over the last few years than it was.
19:02
And about probably about four years ago, I was actually injected with some LNPs myself during COVID.
19:08
So it's something I think has become much more well known.
19:12
Now the issue here is obviously when a liquid nanoparticle is delivered into a tissue, actually detecting protein expression from the LNP or from the constructs within that LNP can be very challenging.
19:26
So what we did or what our collaborator did is actually developed an RNA which was incorporating barcodes on the end of each protein expressed by the RNA which was delivered by the lipid nanoparticles.
19:42
So this was injected in this case into a mouse.
19:47
Marco did mRNA was then transcribed and translated and then the tissues were processed.
19:53
And what we're looking to do identify the mRNA with the highest expression, directly measuring the proteins that were encoded by that messenger RNA.
20:06
What we were able to do is measure the barcodes now because we knew the abundance of the barcodes, how amenable to sequencing those barcodes were.
20:15
That could then act as a proxy for the presence of the proteins encoded by the mRNA construct delivered by the lipid nanoparticles.
20:23
So this has been something that has been developing very quickly.
20:28
We have a number of labs starting to work on it.
20:31
And again, this is an application note that we have around the application, which again, I'm very happy to share with you.
20:38
I'll talk about in more detail if that would be of interest to you.
20:43
So this is really just a very quick tour of the system, how it works, what it does.
20:50
So I see 5 minutes to go.
20:52
So hopefully that's a good time to wrap up.
20:56
So time for a few questions as well.
20:58
So just remains to be seen for me to say.
21:01
Thank you all very much for coming.
21:02
Really appreciate it.
21:03
And if you do have any questions, you're very welcome to ask me now or come to our stand booth 10 as well, where myself and my colleagues, Matthew and Paru will be very pleased to talk to you as well.
21:16
Thanks very much.
